Concise Arguments Against Irreducible Complexity

I'll post these here for quick reference. It is easy for such things to be buried in PZ's many, many posts.

Antecedent versions of the current pathway may have 1) had different functions (the exaptation explanation), 2) had less stringent requirements for function because other physiological functions had less specific demands (the coevolution explanation), or 3) had redundant or alternative paths to the final output of the pathway (the scaffolding explanation). IC, even as defined by the author of the concept, is no obstacle to evolution.

(Emphasis mine.)

Via Pharyngula.